Validating geometry achievement test using one-parameter logistic model: Its rationale and implication for the quality of teacher education

Assessment and examination practices in teacher education influence educators to employ what is common-teacher-made assessments and pre-service teachers to follow them similarly. There is a need to provide standardization and construction of items that fulfill quality indicators, some of which are alignment, fairness, relevance, applicability, objectivity, and validity. This study utilized a one-parameter logistic model to validate a geometry test aimed at assessing lay learning from practice. An 18-item test was developed for 75 pre-service teachers, following an item development cycle, and underwent expert judgment and review. The analysis was conducted using WINSTEPS software. The findings indicate strong reliability in person statistics (separation 2.2, reliability 0.83) and item statistics (reliability coefficient 0.97, separation 6.05). Additionally, the measures explained 56.4% of the raw variation, exceeding the 20% threshold, confirming the test's validity and reliability for assessing geometric achievements. Both the study's process and findings suggest that a conscious approach to item development adheres to a proper cycle, utilizing expert judgment and instrument validation to ensure quality in instruction and teacher competency. This approach also offers teacher educators and pre-service teachers the chance to prepare and validate an instrument aligned with specific content and objectives. These also indicated that assessment and examination play a crucial role in enhancing teacher practice and professional development at teacher education colleges, which is essential for improving teacher education quality. However, further research is recommended due to the small sample size compared to analytic model standards, and the need for employability across various content areas.